Uncertainty-Calibrated Mutation Forecasting Across Folding and Binding Landscapes

Develop a model that outputs both ΔΔG (folding and binding) and per-mutation uncertainty, integrating features from FoldX-like energy terms with MD-derived variability, structural predictions (AF2), and protein–protein affinity predictors. Incorporate kinetic descriptors related to folding rate evolution. Apply this framework to viral proteins (e.g., SARS-CoV-2 RBD) to systematically identify high-affinity or faster-folding mutations not observed in natural variants, while flagging low-confidence regions for targeted assays. This approach addresses the limitation of tools that report point estimates without calibrated uncertainties, leading to brittle decision-making. By learning uncertainty from MD snapshot variability, biochemical context, and model disagreement, it quantifies risk and discovery potential per mutation. It extends prior uncertainty quantification beyond FoldX to multi-task outputs and couples it to modern deep learning structure predictions and protein–protein interactions. It operationalizes the idea of prospecting “unnatural but viable” variants triaged by confidence. The approach promises improved hit rates, reduced wasted experimental cycles, and reveals model blind spots useful for active learning. The impact is a deployable mutation triage engine for protein engineering and pathogen surveillance, enabling safer and more efficient exploration of mutational space.

References:

1. Protein folding rate evolution upon mutations. J. Vila (2023). Biophysical Reviews.
2. Statistical modeling to quantify the uncertainty of FoldX-predicted protein folding and binding stability. Yesol Sapozhnikov, J. S. Patel, F. M. Ytreberg, Craig R. Miller (2023). BMC Bioinformatics.
3. Machine learning methods for protein-protein binding affinity prediction in protein design. Zhongliang Guo, Rui Yamaguchi (2022). Frontiers in Bioinformatics.
4. Highly accurate protein structure prediction with AlphaFold. J. Jumper, Richard Evans, A. Pritzel, Tim Green, Michael Figurnov, O. Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A A Kohl, Andy Ballard, A. Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, J. Adler, T. Back, Stig Petersen, D. Reiman, Ellen Clancy, Michal Zielinski, Martin Steinegger, Michalina Pacholska, Tamas Berghammer, Sebastian Bodenstein, David Silver, O. Vinyals, A. Senior, K. Kavukcuoglu, Pushmeet Kohli, D. Hassabis (2021). Nature.